Character Recognition Using Attention Mechanism for Irregular Text
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing character recognition technologies face inefficiencies when dealing with irregular text, requiring conversion into regular text for recognition, which can be cumbersome and time-consuming.
Innovation Solution
A character recognition method and apparatus that utilizes an attention mechanism for parallel computing to extract image features, calculate attention weight values, and decode characters, thereby simplifying the recognition process and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If irregular text is converted into regular text for recognition, then character recognition can be performed, but the process becomes cumbersome and time-consuming
Solution Approach 1:
The patent segments the image feature vectors into multiple groups and processes them in parallel through multiple processing paths. Each path handles a subset of feature vectors independently, eliminating the need for sequential conversion of irregular text and significantly improving recognition efficiency while reducing time loss.
Solution Approach 2:
The patent implements dynamic parallel processing where the number of processing paths and attention mechanisms adapts based on the input image characteristics. This dynamic approach allows the system to efficiently handle varying text irregularities without fixed conversion procedures, improving both speed and adaptability.
2Productivity
If parallel computing with attention mechanism is used, then operation efficiency is enhanced, but computational complexity increases
Solution Approach 1:
The patent divides the computational task into multiple parallel segments, each processing a subset of image feature vectors through dedicated attention mechanisms. This segmentation distributes computational complexity across multiple independent paths, achieving parallel speedup while keeping individual path complexity manageable.
Solution Approach 2:
The patent applies attention mechanisms selectively to different groups of feature vectors rather than uniformly processing all vectors. This partial application of complex operations to only necessary subsets reduces overall computational complexity while maintaining high efficiency through parallel execution of focused attention calculations.
Data Source
AI summary
A computer device extracts an image feature of an image that includes one or more characters to be recognized. The image feature includes a plurality of image feature vectors. The device uses an attention mechanism to compute and output attention weight values corresponding to the target number of characters, based on the image feature vectors, through parallel computing. Each of the attention weight values corresponds to one or more respective characters and represents an importance of the plurality of image feature vectors for the respective characters. The device obtains at least one character according to the plurality of image feature vectors and the target number of attention weight values. Therefore, in a character recognition process, with recognition based on the foregoing attention mechanism, a character in any shape can be effectively recognized by using a simple procedure, thereby avoiding a cyclic operation process and greatly improving operation efficiency.